The overarching theme of our application is to quantify the impact of cancer cell heterogeneity in tumor growth and treatment resistance. It logically extends results from the previous funding period, pointing to phenotypic heterogeneity as key determinant of progression and invasion. We will consider heterogeneity with respect to phenotypic traits (Proliferation, Motility and Metabolism), in the ICBP-43 breast cancer cell line panel and in drug resistant breast, or radiation responsive lung, cancer cell lines. Trait heterogeneity will be quantified primarily by high-content automated microscopy and image processing. Between cell lines, trait variability will be compared as averages and distribution shapes. Within a cell line, ceil-to-cell variability (presumably non-genetic) will be represented as subpopulations by statistical modeling, e.g., bayesian information criteria and clustering algorithms. To estimate adaptability, we will measure trait variation in response to perturbations mimicking tumor microenvironment conditions. This large dataset (3 traits in >50 lines under >10 perturbations) will be input to mathematical and computational predictive models, tracking the fate of individual cancer cells and the microenvironment in space-time during tumor growth. With the experimental component, this suite of theoretical models forms a Center Backbone deployed towards three Projects. Project 1 will quantify adaptive advantage in cancer progression by incorporating cell trait heterogeneity data into mathematical and computational models that exploit evolution dynamics and game theory concepts. Project 2 will measure impact of trait heterogeneity and fitness cost in the rise of breast cancer resistance to first- and second-line drugs (doxorubicin, hormone therapy and HER2 tyrosine kinase inhibitors). Project 3 will attempt to improve and/or predict outcomes of radiation treatment in lung cancer cell lines by coupling experimentally defined radio-phenotype heterogeneity to predictive models. Hypotheses/predictions from Projects 1-3 will be validated in vitro and in mouse tumors, by iteration loops of experimentation and theory. Finally, we will continue education/outreach efforts, e.g., hands-on cancer modeling workshops, to attract physical and biological scientists, especially the brightest of the new generations.